Dear Readers: (To steal a line from ETS)
Whether you agree with my politics or not, whether you even LIKE politics or not, I want you to read the remainder of this paragraph and the next one before dismissing or ignoring this post. I think that the issue about to be discussed is more important than the specific politics involved and possibly more important than politics in general. It’s a question of trust, honesty, and society.

By now if you haven’t heard about the alleged forgeries that CBS and Dan Rather are bandying about, you must live under a rock. The less you know about this situation, the more important it is for you to read this post. At issue here is the credibility of the mainstream media in general and, in specific, CBS. Now, it’s been long admitted that the media is biased, but there’s a strong difference between “biased” and “unreliable” or “untrustworthy.” The purpose of this post is two-fold. First is to try to hammer down the various sides of this issue for those who only know part of the story or none of it at all. It’s also a personal thing — I want a list of references about the issue myself. This post will be long. I hope that all of you will read it, regardless of your political inclination or affiliations… it’s important to know how reputable our sources of news and information are.

Join me below the fold for the rest of the story.

CBS ran a story on 60 Minutes based on documents, allegedly found from Col. Jerry Killian’s personal file, provided by unnamed sources that CBS claims are unimpeachable. CBS also suggests that other information was received by individuals who worked with Col. Killian. Whether this information provided, again, by unnamed sources is in support of the validity/provenance of the documents themselves or of their contents is unclear. What IS clear, however, is that the case attempting to be made by CBS news and Dan Rather stand solidly on the documents. What I’d like to do, before I delve into the politics of it all, is have a look at the cases made for and against the documents.

First it’s important to have the documents:
1. Memorandum, May 4, 1972
2. Memo to File, May 19, 1972
3. Memorandum For Record, Aug. 1, 1972
4. Memo to File, Aug. 18, 1973

What do they say? Document 1 (D1) is an order by Col. Killian for Lt. Bush to take a physical. Document 2 (D2) is documentation of a conversation between Col. Killian and Lt. Bush. During the conversation, the document alleges that Bush was looking for a way out of drill until after November. It also contains discussion of a physical (presumably the one ordered in D1) and alleges that Lt. Bush said he might not have time to complete it. The last sentence is a mention that Lt. Bush is “talking to someone upstairs.” Document 3 (D3) contains details of the suspension of Lt. Bush for “failure to perform to USAF/TexANG standards and failure to annual physical examination (flight) as ordered.” It also talks of a failed transfer to a different unit. (Presumably one in Alabama) Document 4 (D4) is the one most hotly contested, but seems to say little except that suggestion that someone upstairs is trying to get Col. Killian to falsify records for Lt. Bush. The document has the subject line “CYA.”

As I said before, I’m not yet interested in discussing the content of the documents or the implications thereof… before that can even be considered, we have to decide if the documents are legit.

Proportional Fonts (+)
It’s now widely accepted that Power Line broke the story with suggestions that the documents might not be legit. In the post The Sixty-First Minute (PL1) the author first notes the fact that the memo uses a proportionally spaced font. The second-hand assertion is made that the documents could not be from the date indicated because typewriters could not use proportional spacing. In the first update to the post (PL1u1) the author notes that proportionally spaced fonts did exist on high-end typewriters but suggests that a National Guard post would probably not have such high-end office machinery and also notes that the exact font used has not yet been found.

If we’re going to be fair, the issue of proportionally spaced fonts doesn’t really solve anything. If proportional fonts was NOT available on typewriters at the time, it would be a closed issue. The fact is, however, that there did exist typewriters that could do proportional spacing. In fact, proportional spacing was introduced in the 1940s by IBM (Reference: IBM Archives 1941, SL1). Nevertheless, this is a chink in the armor of the documents, because of the likelihood that an Air National Guard Colonel would have such a high-end machine.

Header Centering (+)
D1 and D3 contain a centered header in a proportionally spaced font. Such a thing is impossible to automatically do on a typewriter and requires a good bit of work, math, and patience to do manually. I do suppose that this could be some sort of letter-head that was produced in bulk for the TANG, but it seems a bit spartan for that. It also seems odd that the type-setter doing the letterhead would use an identical font to the typewriter used to write the memo. You’ll also notice, if you over-lay the two headers and align them, that they’re identical. The likelihood that the three lines would be identically centered on each other in two memos three months apart is extremely low unless: a) a mass-produced letter-head was used or b) Col. Killian (or his typist) had pre-calculated the starting points for each line wrote them down, and used them regularly. In the case of the latter, there’s still a chance that the two would be off by a small amount, which they are not.

Superscripts (+)
D1 and D4 each use a superscript “th” once. In D1 it is after the number “111″ in D4 it is after the number “187.” D1 is a document to be sent out, D4 is a document for file. Notice, first, that “111th” does not contain a superscript in the headers mentioned above. Second, notice that “147th” appears twice in D3 (points 2 and 3) and in neither case is the “th” a superscript. D3, like D1, is a document to be sent out. These are both conspicuous inconsistencies, but neither really prove anything. In the 1970s, how would one go about producing a superscript on a typewriter? Two possible ways. The first is to find a typeset which includes a superscripted “th” key. To the best of my knowledge (and I’ve been reading about this for two days) nobody has asserted that such a typeface existed at all, let alone that it was widespread enough to end up in a Texas Air National Guard typewriter. The second is to do it by hand. This would require the author to stop typing, change the ball of the typewriter to a smaller font, move the paper down about half a line, type the “th”, then reverse the process to go back to typing regularly. The likelihood of this being done for a public memo is debatable (but brought into question by the un-superscripted “th” in D3), the probability of doing all that work for a Cover-Your-Ass memo to file (especially in light of the lack of superscripts in D3) can be nothing but very, very low. Yet this still does not prove a forgery.

Kerning (+)
Kerning, referring to the spacing between two letters, is the thing that allows them to invade each other’s space. To more clearly illustrate this point, you can take a newspaper, draw tiny boxes around each letter that are as small as possible while still containing the entire letter, and then notice that some of the boxes overlap. This is a hallmark of proportionally spaced fonts and not something used at all on fixed-width fonts. Nevertheless, reading the various posts on this topic leaves a great deal of doubt as whether the documents are actually kerned at all, or not. Some claim that it is, some claim that it is not, while some claim that it’s only partially kerned. This is a toss-up that really doesn’t lend much to either side of the debate, but I thought I’d include it here for completeness.

Line Spacing (+)
One thing that the untrained eye would likely fail to notice is the spacing of the lines. Referred to by type-geeks and designers as “leading”, the issue is not obvious, but it amounts to this: the leading used in the documents (13pt?) was virtually unheard of if typewriters at the time. My understanding is that there was only one machine at the time that could’ve been set to produce the spacing in the memos, and that was an IBM Selectric Composer. A source that we’ll reference in just a minute, The Shape of Days, tells us that such machines cost between $3,600 and $4,400 in 1972 which, converted to today’s dollars, is $16,000 to $22,000. The odds that a Texas Air National Guard unit would possess one of these machines is obviously very, very low, but not proof of forgery.

IBM Selectric Composer (+)
If the documents are, indeed, legitimate, all evidence points to the creation of at least one of them with an IBM Selectric Composer. The Shape of Days, in a post titled The IBM Selectric Composer (SOD1)has the definitive piece on this machine as it relates to these documents and comes to two conclusions. First: it seems, on the surface, to be possible that these documents were created using the machine. Second: even with the composer, it would have been a difficult and tiresome task to achieve the superscripting, centering, and linespacing that these documents display. Probability aside, the bottom line is that it is POSSIBLE.

Document Experts (+)
One thing so far lacking from this post is any discussion of what forensic document examiners make of the memos. CBS brought out it’s star witness, a man by the name of Marcel Matley. Matley is a handwriting expert and. According to a post at Wizbang: Rather Defends on CBS News (WB1) Matley tends to ply is wares as an expert witness and write books more often than participate in any sort of scholarly discussion in the field. Moreover, a later post at Wizbang (CBS Star Witness - In his Own Words — hereinafter WB2) notes that in one of Matley’s books he states that one cannot deduce that a document is not a forgery without the original copy, which we’re yet unsure that he has.

More to the point, however, INDC Journal contacted its own document examiner (Are the CBS National Guard Documents Fake? — hereinafter INDC1), Dr. Philip Bouffard who is, from everything I’ve read on him, a very well respected and reputable forensic document examiner. In INDC1, INDC reports Dr. Bouffard’s findings that he believes the document to be fake. The likelihood he placed on it: 90%. Then again, that also doesn’t prove anything… there’s a 10% chance, according to this respected expert, that it’s legit.

Faking It (+)
If a person were to want to create the documents above, how would he or she go about it? On a whim, Little Green Footballs decided to do a little experiment (Another CBS Document Experiment — hereinafter LGF1). Word was opened, in its default settings, and the text of D4 was typed. The match was stunning. The apparent “age” of the document could be faked by making several generations of copies or a few generations of faxes. Now, some have questioned the resolution at which the comparison was made, stating that at higher resolutions the match might be less impressive. This is another possible, yet unconvincing argument for three reasons: first, the quality of the match at the current resolution is very good and the change would be proportional to the change in size. Second, trying to fake the memo in other applications has proven difficult — LGF1 demonstrates with Apple TextEdit. Third, when compared to the copy provided in SOD1, the Microsoft Word version is, without question, a better match. There’s also the small matter that, to my knowledge, none of those questioning the resolution have produced results of the experiment undertaken at higher resolutions, which would be fairly easy to perform.

Facts of the Matter (+)
As the story progresses, the documents are being investigated from more and more angles. Not the least of these is the factual angle of the contents. Killian Memo Has Wrong Deadline, Cites Wrong Regulation from Wizbang (WB2) notes that the regulation cited in point 1 of D1 and point 2 of D3: “&#091In Accordance With&#093 AFM 35-14″ is not the regulation requiring the physical (which is actually 160-1), but the punishment for violating such requirement. Furthermore, 160-1 requirement stated that Lt. Bush would’ve been required to take an annual physical “within 3 months preceding birthday” each year. Bush was born on 7/6/1946, so his physical would need to be after 4/6 and before 7/6. Yet D1 sets a different deadline: 5/14/1972.

On another factual point, Little Green Footballs‘ post Facts: They’re Rather Inconvenient (LGF2) points out that Staudt, referenced in D4 which was allegedly written on 8/1/1973, was honorably discharged on 4/1/1972. How a person would be able to exert the pressure described in D4 16 months after leaving the armed forces is questionable and suspect at best.

Conclusion (+)
You know what? I’m not even going to draw a conclusion here. That’s not really the point. You now have the facts, you have the arguments from both sides presented in the best possible light I could find (and I will correct the record if I find better from EITHER SIDE) so I’m just going to end this post. It will probably grow as the story scampers along, but for now, give it some thought and consider what kind of implications it has if the documents are, indeed, forged and were accepted so quickly by one of the big guys.

If you think I’ve missed something, please email me.


This post will be updated as new developments become available